2024 will be a super bull market in the currency circle, with BTC halved and US dollar interest rates cut. If you are a programmer, quantitative trading robots should learn quickly!
premise, you are familiar with the usage of ssh connection terminal on CVM and the basic python syntax (basic js is also acceptable) This paper uses python quantization robot framework freqtrade to start operation! Freqtrade official documentation There are too many official documents. Please read this article first. For further study, please refer to the official documentation.
1. Prepare the docker environment for CVM
Start with the ubuntu system selected by the CVM, configure the docker environment first, and execute the following commands:
sudo apt update
sudo apt install apt-transport-https ca-certificates curl gnupg-agent software-properties-common
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
sudo apt update
sudo apt install docker-ce
ssh After logging into your server, press the following command in the terminal to start:
it is more convenient to use vscode's ssh remote container development!
mkdir ft_userdata
cd ft_userdata/
curl https://raw.githubusercontent.com/freqtrade/freqtrade/stable/docker-compose.yml -o docker-compose.yml
docker compose pull
docker compose run --rm freqtrade create-userdir --userdir user_data
After execution, the docker-compose.yml file and user_data directory will be created Then initialize the configuration file config.json
docker compose run --rm freqtrade new-config --config user_data/config.json
You will be asked several questions in a row. You can select "enter" as shown in the following log example:
telegram robot creation token and telegram installation, please google search and check
? Do you want to enable Dry-run (simulated trades)? Yes ## 是否使用模拟账户操作，就是模拟交易。设置为false则会用你账户里的真实余额交易
? Please insert your stake currency: USDT ## 货币单位 这里用默认的USDT
? Please insert your stake amount (Number or 'unlimited'): unlimited ## 最大开仓占比，unlimited就是所有金额都会被机器人拿去用。也可以指定如 100，就是只给机器人 100 USDT 去交易
? Please insert max_open_trades (Integer or -1 for unlimited open trades): 3 ## 最大开仓数量，就是可以买几种币。3就是机器人最多只买3种货币。
? Time Have the strategy define timeframe. ## 是否用策略中的时间线，这里用默认的。一般策略中的时间线都是最好的。
? Please insert your display Currency (for reporting): USD ## 机器人展示收益用的单位，可以设置为 CNY，会用人民币表示
? Select exchange okx ## 选择交易所，这里示例选择okx欧易
? Do you want to trade Perpetual Swaps (perpetual futures)? Yes ## 是否启用合约模式。设置false为现货交易。
? Do you want to enable Telegram? Yes ## 启用 telegram 机器人
? Insert Telegram token # telegram 机器人 token
? Insert Telegram chat id # telegram 机器人 id
? Do you want to enable the Rest API (includes FreqUI)? Yes # web管理控制台页面
? Insert Api server Listen Address (0.0.0.0 for docker, otherwise best left untouched) 0.0.0.0 # web管理控制台ip端口
? Insert api-server username freqtrader # web管理控制台账号
? Insert api-server password ****** # web管理控制台密码
The configuration file of the trading robot is the most important, to explain:
{
// 交易模式，这里有三个选项，spot (默认，现货交易)，future(期货交易，可以进行做空交易)， margin（当前不可用）
"trading_mode": "spot",
// 当交易模式不是spot时，需要设置这里，当交易模式为future时，要设置为isolated， 当交易模式为margin时设置为cross（暂时不可用）
// "margin_mode": "isolated",
// 最大的单数
"max_open_trades": 5,
// 用于交易的加密货币
"stake_currency": "USDT",
// 机器人可用的金额，可以开启多个机器人用于交易，平横策略间的收益差异
"stake_amount": 200,
// 允许机器人交易的账户总余额的比率
"tradable_balance_ratio": 1,
// 从虚拟币转换为哪种法定货币，这里时美金，如果转换报错，可以不设置
// "fiat_display_currency": "USD",
// 运行模式，虚拟运行，还是真实运行，true是虚拟运行
"dry_run": true,
// 要使用的时间帧 1m，5m, 15m, 30m, 1h ...,这个歌配置可以在这里缺省，并在策略文件进行再次配置，也就是策略文件的配置强于这里
"timeframe": "3m",
// 虚拟运行的总金额
"dry_run_wallet": 1000,
// 在退出时取消未结订单
"cancel_open_orders_on_exit": true,
// 必须的配置
"unfilledtimeout": {
// 只要有信号，机器人将等待未完成的挂单完成多长时间（以分钟或秒为单位），之后订单将被取消并以当前（新）价格重复。
"entry": 10,
// 只要有信号，机器人将等待未履行的退出订单完成多长时间（以分钟或秒为单位），之后订单将被取消并以当前（新）价格重复
"exit": 30
},
// 交易所设置
"exchange": {
"name": "okx",
// API key
"key": "",
// API secret
"secret": "",
// ccxt的一些配置，在国外一般用不到，在国内可以在这里配置代理
"ccxt_config": {},
"ccxt_async_config": {},
// 交易对白名单
// 1INCH/USDT是现货交易
// 1INCH/USDT:USDT是期货交易
// 在这里可以使用通配符进行批量的配置，不必列出所有交易对
// 查看交易所支持的交易对，命令：freqtrade list-markets -c config.json 或者是 freqtrade list-pairs -c config.json
"pair_whitelist": [
"1INCH/USDT:USDT",
"ALGO/USDT:USDT"
],
// 黑名单
"pair_blacklist": [
// 黑名单通配符设置，白名单同理
"*/BNB"
]
},
"entry_pricing": {
// 必须配置，选择机器人应查看价差的一侧以获得进入率。默认same 还有 ask, bid, other
"price_side": "same",
// 允许使用订单簿输入中的费率进行输入。缺省值为true
"use_order_book": true,
// 机器人将使用订单簿“price_side”中的前 N 个汇率来进入交易。即，值为 2 将允许机器人选择Order Book Entry中的第二个条目。
"order_book_top": 1,
// 必须配置，插值投标价格。
"price_last_balance": 0.0,
"check_depth_of_market": {
// 如果Order Book中的买单和卖单有差异，则不要进入。
"enabled": false,
// 订单簿中的买单与卖单的差额比例。值低于 1 表示卖单规模较大，而值大于 1 表示买单规模较大
"bids_to_ask_delta": 1
}
},
"exit_pricing": {
// 选择机器人应查看的价差一侧以获得退出率。默认same
"price_side": "other",
"use_order_book": true,
"order_book_top": 1
},
// 定义要使用的一个或多个配对列表。
"pairlists": [
// 这里的方法有很多， [点击查看详情](https://www.freqtrade.io/en/stable/plugins/#pairlists-and-pairlist-handlers)
// 大部分默认情况下都是使用StaticPairList
{
"method": "StaticPairList"
}
],
// 后面还可以添加frequi的配置，以及telegrambot的配置
"bot_name": "",
"force_entry_enable": true,
"initial_state": "running",
"internals": {
"process_throttle_secs": 5
}
}
In config configuration, the most important ones are dryrun and tradingmode. Exchange needs to be matched with the API key and secret of the exchange, okx can create an apiKey in the API on the app. Put my own config file here, which is the type of contract transaction:
{
"tradable_balance_ratio": 0.99,
"fiat_display_currency": "CNY",
"stake_amount": "unlimited",
"stake_currency": "USDT",
"amend_last_stake_amount": true,
"dry_run_wallet": 1600,
"max_open_trades": 10,
"dry_run": false,
"cancel_open_orders_on_exit": false,
"trading_mode": "futures",
"margin_mode": "isolated",
"unfilledtimeout": {
"entry": 10,
"exit": 10,
"exit_timeout_count": 0,
"unit": "minutes"
},
"entry_pricing": {
"price_side": "other",
"use_order_book": true,
"order_book_top": 1,
"price_last_balance": 0.0,
"check_depth_of_market": {
"enabled": false,
"bids_to_ask_delta": 1
}
},
"exit_pricing": {
"price_side": "other",
"use_order_book": true,
"order_book_top": 1
},
"exchange": {
"name": "okx",
"key": "",
"secret": "",
"password": "",
"ccxt_config": {},
"ccxt_async_config": {},
"pair_blacklist": [
"(BTCUSDT_.*|ETHUSDT_.*)",
"(GT|HT)/.*",
"(WBTC|BSV|BTCDOM|DEFI)/.*",
".*(_PREMIUM|BEAR|BULL|DOWN|HALF|HEDGE|UP|[1235][SL])/.*",
"(AUD|BRZ|CAD|CHF|EUR|GBP|HKD|IDRT|JPY|NGN|RUB|SGD|TRY|UAH|USD|ZAR|UST)/.*",
"(BUSD|CUSDT|DAI|PAX|PAXG|SUSD|TUSD|USDC|USDT|VAI|USDN)/.*",
"(ACM|AFA|ALA|ALL|APL|ASR|ATM|BAR|CAI|CITY|FOR|GOZ|IBFK|LEG|LOCK-1|NAVI|NOV|OG|PFL|PSG|ROUSH|STV|TH|TRA|UCH|UFC|YBO)/.*"
]
},
"pairlists": [
{
"method": "VolumePairList",
"number_assets": 20,
"sort_key": "quoteVolume",
"min_value": 0,
"refresh_period": 86400,
"lookback_days": 1
},
{
"method": "AgeFilter",
"min_days_listed": 10
},
{
"method": "PrecisionFilter"
},
{
"method": "PriceFilter",
"low_price_ratio": 0.01
},
{
"method": "SpreadFilter",
"max_spread_ratio": 0.002
},
{
"method": "RangeStabilityFilter",
"lookback_days": 3,
"min_rate_of_change": 0.02,
"refresh_period": 1440
},
{
"method": "ShuffleFilter",
"seed": 42
}
],
"telegram": {
"enabled": true,
"token": "",
"chat_id": "",
"keyboard": [
[
"/daily",
"/status table",
"/count"
],
[
"/profit",
"/performance",
"/balance"
],
[
"/status",
"/show_config",
"/whitelist"
],
[
"/logs",
"/reload_config",
"/help"
]
]
},
"edge": {
"enabled": false,
"process_throttle_secs": 3600,
"calculate_since_number_of_days": 14,
"allowed_risk": 0.01,
"stoploss_range_min": -0.01,
"stoploss_range_max": -0.1,
"stoploss_range_step": -0.01,
"minimum_winrate": 0.60,
"minimum_expectancy": 0.20,
"min_trade_number": 10,
"max_trade_duration_minute": 1440,
"remove_pumps": false
},
"api_server": {
"enabled": false,
"listen_ip_address": "127.0.0.1",
"listen_port": 8080,
"verbosity": "error",
"enable_openapi": false,
"jwt_secret_key": "",
"ws_token": "",
"CORS_origins": [],
"username": "",
"password": ""
},
"bot_name": "freqtrade",
"initial_state": "running",
"force_entry_enable": true,
"internals": {
"process_throttle_secs": 5
}
}
In the user_data/strategies directory, you will be given a default policy file with a brief description:
mplifying code
class SampleStrategy(IStrategy):
INTERFACE_VERSION = 3 # 版本为3的才可以做空
can_short: bool = False # 是否可以做空
# 止盈
minimal_roi = {
"60": 0.01,
"30": 0.02,
"0": 0.04
}
# 止损
stoploss = -0.10
# 移动止损
trailing_stop = False
# 用哪个时间线的k线
timeframe = '5m'
# 设置指标
def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
return dataframe
# 入场
def populate_entry_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
dataframe.loc[
(
(qtpylib.crossed_above(dataframe['rsi'], self.buy_rsi.value))
),
'enter_long'] = 1 # 开仓信号，做多
dataframe.loc[
(
(qtpylib.crossed_above(dataframe['rsi'], self.short_rsi.value))
),
'enter_short'] = 1 # 开仓信号，做空
return dataframe
# 出场
def populate_exit_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
dataframe.loc[
(
(qtpylib.crossed_above(dataframe['rsi'], self.sell_rsi.value)) # rsi穿过sell_rsi
),
'exit_long'] = 1 # 平仓信号，平多
dataframe.loc[
(
(qtpylib.crossed_above(dataframe['rsi'], self.exit_short_rsi.value)) # rsi穿过exit_short_rsi
),
'exit_short'] = 1 # 平仓信号，平空
return dataframe
This is the strategy of the example, it may not be easy to use, there are many websites in the community that can find the strategy and copy it directly to use it.
Open the docker.compose.yml file:
---
version: '3'
services:
freqtrade:
image: freqtradeorg/freqtrade:stable # 官方镜像
restart: unless-stopped
container_name: freqtrade
volumes:
- "./user_data:/freqtrade/user_data"
ports:
- "127.0.0.1:8080:8080"
command: >
trade
--logfile /freqtrade/user_data/logs/freqtrade.log # 输出日志的文件
--db-url sqlite:////freqtrade/user_data/tradesv3.sqlite # 数据库
--config /freqtrade/user_data/config.json # 机器人配置文件
--strategy SmartTA # 策略文件class名
The focus here is on the-- strategy parameter, followed by the class name in the policy code, not the file name Terminal execution:
docker compose up -d
Boot, and then you can see the startup log in the user_data/logs/freqtrade.log file. The telegram robot will also print out logs.
2 strategy websites are recommended:
Finally, let's talk about the back test. Freqtrade provides the back test function. I can only say that it can only verify whether the code logic of the policy file has bug. The income gap between the return test and the normal test is still quite obvious. It is recommended that the actual environment dry-run is stable for a period of time before opening a real account.